Master’s Degree Awarded to Ms. Hiba Al-Ba’adani in Oral and Maxillofacial Surgery
Ms. Hiba Abdo Mohammed Za’atan Al-Ba’adani was awarded a Master’s Degree in Oral and Maxillofacial Surgery for her thesis titled: The Oral Biopsy versus Artificial Intelligence in Detection of Dental Cysts, Abscesses, and Tumors, which was submitted to the Faculty of Dentistry – Sana’a University. The MA defense was held on Saturday, March 14, 2026.
The MA Viva-voce Committee, which was formed based on a resolution issued by the Graduate Studies and Scientific Research Council, consisted of the following:
| # | Committee Members | Designation | Position |
| 1 | Assoc. Prof. Ahmed Saleh Yahya Al-Khatri | External Examiner | Chair |
| 2 | Assoc. Prof. Sam Abdulkarim Mohammed Da’er | Main Supervisor | Member |
| 3 | Dr. Abdullah Farhan Al-Sharabi | Internal Examiner | Member |
The thesis aimed to:
- Develop and evaluate two separate deep learning models for the differential diagnosis of periapical lesions.
- Build a classification model using EfficientNetB3 architecture to categorize lesions into abscesses, cysts, and tumors.
- Develop a second model for lesion detection and segmentation using the YOLO
- Explore an assistive approach to reduce reliance on oral biopsy as an invasive and costly procedure, by utilizing artificial intelligence for early and accurate diagnosis of oral
The study yielded several key findings summarized as follows:
- The classification model achieved a high accuracy of 96.40%, with good performance in identifying abscesses (83.3%) and cysts (61.9%), but was unable to detect tumors due to limited available data.
- The detection model using YOLO achieved sensitivities of 65% for abscesses, 5% for cysts, and 30% for tumors, highlighting the need for larger and more balanced datasets to improve the performance of the models.
- The study confirmed that data size and balance are critical factors in the effectiveness of medical AI systems, particularly in rare tumor cases.
In light of these findings, the researcher recommended the following:
- Enhancing interactive content by developing more engaging and user-friendly tools, such Developing larger and more balanced datasets, especially for tumor cases.
- Expanding the use of artificial intelligence applications in the diagnosis of oral and dental diseases.
- Conducting further studies to improve model performance and enable clinical
- Enhancing interdisciplinary collaboration between medical and technical fields.
The defense session was attended by a number of academics, researchers, students, colleagues, and the researcher’s family.
